A significant graft versus leukemia (GVL) immune response, mediated by donor lymphocytes after allogeneic bone marrow transplantation (BMT), contributes to the cure of 40-50 percent of adult patients with myeloid leukemias. However, allogeneic BMT is only available for 25 percent of all patients due to patient age and donor availability. Therefore, the overall objective of this proposal is to develop strategies for the generation of antigen-specific antileukemia immune responses for the treatment of myeloid leukemias that can be applied to more patients. Proteinase 3 (Pr3) is a 26kd myeloid-restricted azurophil granule protein that is normally maximally expressed only in developing promyelocytes and is an antigen for the antineutrophil cytoplasmic antibody (ANCA) in patients with Wegener's granulomatosis. Pr3 is overexpressed by 3 to 6 fold in 75 percent of chronic myeloid leukemia (CML) and 50 percent of acute myeloid leukemia (AML) cells. An HLA-A2-restricted 9 amino acid peptide derived from Pr3, PR1, has been shown to be a target epitope of cytotoxic lymphocytes (CTL) that preferentially lyse myeloid leukemia over normal bone marrow progenitors in vitro. The amount of CTL lysis correlates with Pr3 overexpression in the target cells. Detection of immunity to PR1 in normal donors and some patients with myeloid leukemia strongly implies that patient's own leukemia cells can act as a source of immunizing protein and begs the issue of whether existent immune responses to PR1 play any role in slowing leukemia progression and whether boosting of immune responses can offer any therapeutic benefit. The current NEW INVESTIGATOR proposal will assess the clinical importance of immune responses to PR1 peptide, evaluate a PR1 peptide-based tumor vaccine in patients with myeloid leukemia, and since Pr1 is the first tissue-restricted epitope identified, will investigate other unique HLA-A2-associated Pr3 peptides as leukemic-specific epitopes for CTL. To investigate the clinical significance of immune responses to PR1, CTL precursor (CTLP) frequency against PR1 will be determined in patients with CML, AML and their normal marrow donors before and after BMT using limiting dilution analysis and by using a fluorescence-tagged PR1-tetramer to directly label PR1-specific CTL. Pr3 expression in leukemia and the surface phenotype of leukemia and PR1-specific CTL will be examined and correlated to CTLP frequency and clinical outcomes. To investigate whether an immune response to PR1 can be boosted, a phase I/II study (NCI study number T98-0017, Molldrem P.I.) of PR1 peptide-based vaccine with incomplete Freund's adjuvant (IFA) will be conducted in patients with CML, AML, and myelodysplastic syndrome. The University of Texas M. D. Anderson Cancer Center has approved the investigation, and NCI will supply GMP quality PR1 peptide and IFA. Immune responses to the vaccine will be determined by LDA of CTLP and by analyzing the number of PR1-specific CTL with the fluorescent PR1-tetramer and will be correlated to clinical response and percent of leukemia blasts in bone marrow. Lastly, other synthetic HLA-A2-associated peptides from Pr3 will be investigated for their potential to elicit peptide-specific CTL from donor PBMC in vitro and then tested for ability to lyse myeloid leukemia.